Subject: Re: How Should We *Motivate* Students in Intro Stat?

     To: edstat-l@jse.stat.ncsu.edu
         sci.stat.edu Usenet newsgroup

   From: Donald Macnaughton <donmac@matstat.com>
                   (formerly donmac@hookup.net)

   Date: Wednesday December 18, 1996 21:14 EDT

     Cc: Dennis Roberts <dmr@email.psu.edu>

On December 17, 1996, Dennis Roberts wrote

> i was not implying that there is 0 accuracy ... it is just when
> a term like "accurate" is bandied about ... one gets the im-
> pression that the error is relatively small. but .. for most
> predictors and criteria ... when predicting human behavior var-
> iables ... the errors variance is normally greater ... many
> times MUCH greater ... than the predictable variance.  that's
> ALL i meant ...

Dennis makes an important point:  When we show students how the 
field of statistics helps researchers to make "accurate" predic-
tions, we must explain what we mean by the word "accurate".  We 
must also inform students that statistical methods can usually 
provide us with an estimate of the size of the *errors* in the 
predictions that (on the basis of the available research data) 
the procedure will make.  

(Furthermore, as is well known, we must explain to students the 
concept of the *underlying assumptions* of statistical procedures 
because an estimate of size of the prediction errors will be re-
liable only if the underlying assumptions of the associated sta-
tistical procedure are adequately satisfied.)

I give an approach to discussing these matters with students in a 
paper for students (1996, sections 11.2, 9.4, 9.10, and 11.4).

Don Macnaughton


REFERENCE
Macnaughton, D. B. (1996), "The Entity-Property-Relationship Ap-
   proach to Statistics:  An Introduction for Students."  Avail-
   able at http://www.matstat.com/teach/